Abstract:In real-world applications, robotic systems collect vast amounts of new data from ever-changing environments over time. They need to continually interact and learn new knowledge from the external world to adapt to the environment. Particularly, lifelong object recognition in an online and interactive manner is a crucial and fundamental capability for robotic systems. To meet this realistic demand, in this article, we propose an online active continual learning (OACL) framework for robotic lifelong object recognition, in the scenario of both classes and domains changing with dynamic environments. First, to reduce the labeling cost as much as possible while maximizing the performance, a new online active learning (OAL) strategy is designed by taking both the uncertainty and diversity of samples into account to protect the information volume and distribution of data. In addition, to prevent catastrophic forgetting and reduce memory costs, a novel online continual learning (OCL) algorithm is proposed based on the deep feature semantic augmentation and a new loss-based deep model and replay buffer update, which can mitigate the class imbalance between the old and new classes and alleviate confusion between two similar classes. Moreover, the mistake bound of the proposed method is analyzed in theory. OACL allows robots to select the most representative new samples to query labels and continually learn new objects and new variants of previously learned objects from a nonindependent and identically distributed (i.i.d.) data stream without catastrophic forgetting. Extensive experiments conducted on real lifelong robotic vision datasets demonstrate that our algorithm, even trained with fewer labeled samples and replay exemplars, can achieve state-of-the-art performance on OCL tasks.

Object Recognition Consistency in Regression for Active Detection

Consistency-based Active Learning for Object Detection

An Over-Regression Suppression Method to Discriminate Occluded Objects of Same Category

Localization-Aware Active Learning for Object Detection

Hierarchical Regression and Classification for Accurate Object Detection

Accurate Object Detection With Location Relaxation And Regionlets Re-Localization

Decouple and Align Classification and Regression in One-Stage Object Detection

Reconcile Prediction Consistency for Balanced Object Detection

Region-level Active Detector Learning

Online Active Continual Learning for Robotic Lifelong Object Recognition

Uncertainty Aware Active Learning for Reconfiguration of Pre-trained Deep Object-Detection Networks for New Target Domains

Exploring Residual and Spatial Consistency for Object Detection

Hybrid Classification-Regression Adaptive Loss for Dense Object Detection

Recalibrating Features and Regression for Oriented Object Detection.

A Task-Risk Consistency Object Detection Framework Based on Deep Reinforcement Learning

Classification Committee for Active Deep Object Detection

Stochastic Context Consistency Reasoning for Domain Adaptive Object Detection

DeLR: Active Learning for Detection with Decoupled Localization and Recognition Query

Deep Active Learning for Remote Sensing Object Detection

Learning Adversarially Robust Object Detector with Consistency Regularization in Remote Sensing Images

OL-Reg: Registration of Image and Sparse LiDAR Point Cloud with Object-Level Dense Correspondences